Traditionally, vehicle routing is optimized for travel time or distance. However, there are contexts where routes that prioritize safety over other criteria are needed. This disclosure describes techniques that use publicly-available vehicle crash data and machine learning models trained on roadway characteristics, driver behavior, environmental factors, etc. to predict the likelihood of an accident on a given route. The historical data and the predictions made by the model are used as signals to rank recommended routes in a navigation system.

Creative Commons License

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.